INDEX

Explanations

occurrences of the word "in"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RenderAtEndOf

-0.43

//

-0.42

MergeFrom

-0.39

nościo

-0.37

 Frü

-0.35

gehen

-0.35

 surla

-0.34

 Bergmann

-0.34

-0.33

 come

-0.33

POSITIVE LOGITS

 maint

1.41

 maintenance

0.91

Maintenance

0.91

 mainten

0.88

maintenance

0.88

 Maintenance

0.87

 MAINTENANCE

0.82

Maintain

0.81

Maintaining

0.80

 Maintain

0.78

Activations Density 0.000%

occurrences of the word "in"

No Comments

No Known Activations